Many years ago, Christoph Herwig, a bioprocessing and datascience expert, became fascinated by a fundamental question: how to use data or data analysis to get a robust (bio)process? That early curiosity grew into his mission ‘Affordable drugs for the world!’. Today, he applies data science to the bioprocess life cycle, using digital models to help companies in pharma and food optimise their processes, reduce variability, and overcome long-standing challenges—especially in the increasingly important realm of single-use technologies.
Herwig already worked with single-use equipment 20 years ago and saw a lot of things go wrong. “For example rupture of bags, no proper mixing, suboptimal processing conditions and a lot of variability between batches.” Around that time the principles of ‘Quality by Design’ (QbD), emerged which is a systematic approach to developing and controlling biopharmaceutical manufacturing processes. Instead of relying mainly on end-product testing to confirm quality, QbD builds quality into the process from the very beginning. Herwig became very interested in assessing how batch variability can be reduced by the concepts introduced by QbD.
Impact
From 2008 to 2023 Herwig worked as a professor in bioprocess engineering at the Vienna University of Technology. He focused a lot on methods for data analysis. “This has generated around 400 scientific papers projecting data science to bioprocessing during my time as a professor. Publications are of course nice, but it is much more rewarding when you make impact by turning those ideas into practice. For example, improving a process, solving a problem, or benefitting society. Then the finding in the paper gets value.”
Companies
To make more impact he started a company in 2013 in data science, which developed data science tools like products in consultancy as well as in software. In 2020 the company was bought by Körber Pharma (pioneering data science software solutions for the biopharma life cycle) and he currently is senior scientific advisor for Körber. He also has started different small companies. An example is Fermify that produces casein (milk protein) —using microorganisms instead of cows for the production of animal-free cheese and dairy alternatives. “Although I am no longer employed by the university I am still very much connected to the scientific world and innovation strongly happens outside of universities. I apply data science principles to develop complete applications. You could say that data science is the new oil.”
Digital twin
“What we do is analyse data, establish the link between the input and the output of the process and turn this into a digital model of the process”, explains Herwig. “If you apply the model in the real-time context, it’s called a digital twin, which is actually a virtual replica of the process.”
A digital twin simulates how cells, media, and process parameters interact to produce the product with the right quality as well as with optimum economy and sustainability. It lets you test and optimise conditions in advance and then adjust in real time to handle variability. It also helps ensure consistent end quality across processes process steps and allows decentralized manufacturing.”
“Furthermore it is very well-suited for single-use processes, where limited sensors make control difficult. Even with minimal inline data, a digital twin can infer missing information from what is measured, reducing the need for extensive testing and saving time and money. More efficient biopharmaceutical processes can ultimately lower drug costs worldwide”, a goal that Herwig considers central to his mission.
Christoph Herwig will be giving a lecture at Single-Use Event Basel 2026 about triggering enhanced process robustness and predictability in SUT by AI-based digital twins.
Single-Use Event Basel 2026 is the ultimate networking event for professionals in bioprocessing, biotechnology and (bio)pharmaceutical manufacturing, with a focus on single use. Taking place on March 19, 2026, at Messe Basel, the event offers a full day of innovation, insights and interaction.